The objective of this study is to identify mechanisms that explain why some adolescents fare worse than others in response to family and community violence. The study hypothesizes that violence exposure erodes family processes and adolescents' daily experiences, and that these proximal variables account for compromised adolescent functioning. The study provides in-depth perspectives of family processes and ongoing daily experiences through behavioral samples of family and peer interactions, and through 14 days of daily diary data. The study also hypothesizes that violence exposure leads to dysregulated biological stress of the hypothalamic-pituitary-adrenal (HPA) axis (measured through cortisol), and that violence-exposed youth are primed for HPA activation in response to negative family interactions. This study extends an ongoing investigation involving three previous assessments of marital violence, parent-child aggression, and community violence in an ethnically diverse sample of two-parent families with a child age 9 or 10 when the study began. The proposed project adds three more assessments of these exposure variables, and extends the study from pre-adolescence through late adolescence. Six waves of data collection allow for the study of cascading effects across time, with feedback loops from negative family processes, adolescent daily experiences and adolescent adjustment contributing in a bi-directional fashion to the likelihood of violence. Repeating, detailed assessments of family and community violence allow for testing whether multiple types of violence, violence at an earlier age, and more persistent violence contribute to more pervasive and deleterious outcomes. Adolescence is targeted because of the importance of understanding how the risks associated with violence exposure coincide with the normal challenges at this time, and because adolescents' abilities to successfully negotiate educational, social, and health domains are highly salient to their subsequent physical and psychological adjustment as young adults. The study has multiple reporters (adolescent, mother, father, peer), uses multiples types of data collection, and measures multiple developmental outcomes (behavioral and psychological functioning; aggression toward others; academic achievement; and high-risk, health-compromising behaviors). Through growth curve models estimated using partial least squares latent variable analysis, we examine multiple pathways through which violence exposure influences adolescent outcomes. The focus on proximal, modifiable variables ultimately can be used to inform intervention and prevention programs for adolescents exposed to violence.